目前文本情感分类主要是针对文本情感的倾向性分析,主要研究正负情感的分类。主要研究短文本领域的文本多情感分析工作,通过建立多项式贝叶斯分类模型以及结合状态空间,构建并训练整个分析模型。实验结果表明,文本的情感分类在实际结果中具有一定的合理性,可以通过算法的分析得出多种情感的分析结果。
Nowadays,text sentiment classification is mainly for text sentiment propensity analysis,positive and negative emotion classification.Studies the text short text in the field of multi-sentiment analysis work,through the establishment of a multinomial model of Bayes classifier and the state-space representation,to build and train the entire model.Experimental results show that emotional text classification has certain rationality in actual results,analysis results can be obtained through a variety of emotional analysis algorithm.